occamy-1.0-MLX-5bit / validation.json
Eang's picture
Release native MLX 5bit with artifact, HTTP and paired BF16 quality checks
b3a65ec verified
Raw History Blame Contribute Delete
490 kB
{
"key": "5bit",
"mode": "affine",
"bits": 5,
"group_size": 64,
"scope": "Complete artifact and native dequantized-value checks, plus eight authored cached greedy fixtures; not a broad benchmark.",
"backend": "Linux native MLX CUDA 12 on NVIDIA B200",
"mac_acceptance": "pending",
"files": [
{
"file": "model-00001-of-00005.safetensors",
"bytes": 5362048467,
"sha256": "88e2aced3d2273877135f295a2d11cbd48049956613dad4065fb7bea1eb5db00"
},
{
"file": "model-00002-of-00005.safetensors",
"bytes": 5215907743,
"sha256": "42727f19852b6f7cf33f3dc714a09f7b9c85d9a90fa32fe7d9937e092dfab7da"
},
{
"file": "model-00003-of-00005.safetensors",
"bytes": 5207519082,
"sha256": "0a5b21f2449651d4079aaf747651bdbacfb3b68bbff744e029bca997a8cde77d"
},
{
"file": "model-00004-of-00005.safetensors",
"bytes": 5203037976,
"sha256": "4010416518fbd293a7744289f8e8edeba8f0c0b9170933caee3d8ff768ba30cb"
},
{
"file": "model-00005-of-00005.safetensors",
"bytes": 2850310361,
"sha256": "9442f609345540abbe0f0ecfa541fe19d42c603598dfa775ee399ed63c4ede1d"
}
],
"tensor_checks": [
{
"name": "language_model.model.layers.8.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.embed_tokens.scales",
"shape": [
248320,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.embed_tokens.biases",
"shape": [
248320,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.embed_tokens.weight",
"shape": [
248320,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.0.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.7.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.2.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.6.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.6.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.1.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.3.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.5.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.0.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.3.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.4.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.2.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.7.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.5.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.8.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.11.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.10.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.9.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.10.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.8.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.16.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.12.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.14.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.12.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.15.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.14.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.13.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.15.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.19.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.23.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.24.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.21.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.18.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.21.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.24.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.20.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.18.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.17.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.22.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.25.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.25.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.19.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.23.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.26.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.26.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.29.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.27.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.33.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.28.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.29.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.31.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.33.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.31.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.27.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.30.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.34.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.28.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.32.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.norm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.v_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.q_proj.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.v_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.q_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.o_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.k_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.self_attn.k_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.k_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.q_proj.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.lm_head.scales",
"shape": [
248320,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.self_attn.v_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.q_proj.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_z.biases",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.self_attn.o_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.out_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.gate.scales",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.lm_head.biases",
"shape": [
248320,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_b.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.self_attn.o_proj.weight",
"shape": [
2048,
640
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.35.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.35.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.36.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.down_proj.biases",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert_gate.biases",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_a.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.up_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.conv1d.weight",
"shape": [
8192,
4,
1
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.up_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.gate_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.lm_head.weight",
"shape": [
248320,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.up_proj.scales",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.up_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_qkv.weight",
"shape": [
8192,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.out_proj.biases",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.norm.weight",
"shape": [
128
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.gate.biases",
"shape": [
256,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.self_attn.k_norm.weight",
"shape": [
256
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.gate.weight",
"shape": [
256,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.down_proj.weight",
"shape": [
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.shared_expert_gate.weight",
"shape": [
1,
512
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.post_attention_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.down_proj.scales",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.shared_expert_gate.scales",
"shape": [
1,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.shared_expert.down_proj.biases",
"shape": [
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.in_proj_a.weight",
"shape": [
32,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.down_proj.scales",
"shape": [
256,
2048,
8
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.down_proj.weight",
"shape": [
256,
2048,
80
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.gate_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.up_proj.biases",
"shape": [
256,
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.gate_proj.scales",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.mlp.switch_mlp.up_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.input_layernorm.weight",
"shape": [
2048
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.39.mlp.shared_expert.gate_proj.weight",
"shape": [
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.37.linear_attn.A_log",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.36.mlp.switch_mlp.gate_proj.weight",
"shape": [
256,
512,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.linear_attn.dt_bias",
"shape": [
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_b.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_a.biases",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_b.scales",
"shape": [
32,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_qkv.biases",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_qkv.scales",
"shape": [
8192,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_z.scales",
"shape": [
4096,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.mlp.shared_expert.gate_proj.biases",
"shape": [
512,
32
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
},
{
"name": "language_model.model.layers.38.linear_attn.in_proj_z.weight",
"shape": [
4096,
320
],
"dtype": "mlx.core.uint32"
},
{
"name": "language_model.model.layers.38.linear_attn.out_proj.scales",
"shape": [
2048,
64
],
"dtype": "mlx.core.bfloat16",
"all_stored_values_finite": true
}
],
"generation": [
{
"id": "text_en",
"messages": [
{
"role": "user",
"content": "Explain in two short sentences why dividing a job into smaller tasks can help a team."
}
],
"content": "Dividing a job into smaller tasks allows team members to specialize and work on specific components simultaneously, which significantly boosts overall efficiency. This approach also clarifies individual responsibilities, making it easier to track progress and manage deadlines.",
"expected": null,
"passed": true,
"finish_reason": "stop",
"tokens": 44,
"full_vocab_finite_logit_checks": 45,
"seconds": 29.96307349205017
},
{
"id": "text_zh",
"messages": [
{
"role": "user",
"content": "请用两句话说明为什么重要文件需要备份。"
}
],
"content": "备份重要文件能有效防止因硬件故障、病毒攻击或人为误删等意外情况导致的数据永久丢失。它是确保信息在灾难发生后能够快速恢复并维持业务连续性的最后一道安全防线。",
"expected": null,
"passed": true,
"finish_reason": "stop",
"tokens": 42,
"full_vocab_finite_logit_checks": 43,
"seconds": 3.2290306091308594
},
{
"id": "math_1",
"messages": [
{
"role": "user",
"content": "Return only the integer: 37 + 58."
}
],
"content": "95",
"expected": "95",
"passed": true,
"finish_reason": "stop",
"tokens": 3,
"full_vocab_finite_logit_checks": 4,
"seconds": 0.9851634502410889
},
{
"id": "math_2",
"messages": [
{
"role": "user",
"content": "Return only the integer: 144 divided by 12."
}
],
"content": "12",
"expected": "12",
"passed": true,
"finish_reason": "stop",
"tokens": 3,
"full_vocab_finite_logit_checks": 4,
"seconds": 1.2618868350982666
},
{
"id": "json_1",
"messages": [
{
"role": "user",
"content": "Return only valid JSON with no markdown. The object must be exactly {\"city\":\"Paris\",\"count\":3}."
}
],
"content": "{\"city\":\"Paris\",\"count\":3}",
"expected": {
"city": "Paris",
"count": 3
},
"passed": true,
"finish_reason": "stop",
"tokens": 10,
"full_vocab_finite_logit_checks": 11,
"seconds": 14.993015766143799
},
{
"id": "json_2",
"messages": [
{
"role": "user",
"content": "Return only valid JSON with no markdown: an array of the even integers from 2 through 8, in increasing order."
}
],
"content": "[2, 4, 6, 8]",
"expected": [
2,
4,
6,
8
],
"passed": true,
"finish_reason": "stop",
"tokens": 13,
"full_vocab_finite_logit_checks": 14,
"seconds": 2.6727020740509033
},
{
"id": "memory_en",
"messages": [
{
"role": "user",
"content": "Remember the code word: CORAL-27."
},
{
"role": "assistant",
"content": "The code word is CORAL-27."
},
{
"role": "user",
"content": "Return only the code word I gave you."
}
],
"content": "CORAL-27",
"expected": "CORAL-27",
"passed": true,
"finish_reason": "stop",
"tokens": 6,
"full_vocab_finite_logit_checks": 7,
"seconds": 0.9007015228271484
},
{
"id": "memory_zh",
"messages": [
{
"role": "user",
"content": "记住:仓库编号是海风42。"
},
{
"role": "assistant",
"content": "仓库编号是海风42。"
},
{
"role": "user",
"content": "只回答刚才的仓库编号。"
}
],
"content": "海风42",
"expected": "海风42",
"passed": true,
"finish_reason": "stop",
"tokens": 5,
"full_vocab_finite_logit_checks": 6,
"seconds": 0.8554456233978271
}
],
"all_stored_floating_values_finite": true,
"weight_bytes": 23838823629,
"strict_stock_reload": true,
"quantized_modules": [
{
"module": "language_model.lm_head",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 248320,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.embed_tokens",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 248320,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.0.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.1.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.10.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.11.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.12.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.13.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.14.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.15.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.16.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.17.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.18.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.19.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.2.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.20.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.21.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.22.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.23.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.24.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.25.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.26.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.27.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.28.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.29.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.3.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.30.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.31.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.32.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.33.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.34.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.35.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.36.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.37.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.38.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.39.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.4.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.5.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.6.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.self_attn.k_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.self_attn.o_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.self_attn.q_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.7.self_attn.v_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.8.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.linear_attn.in_proj_a",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.linear_attn.in_proj_b",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 32,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.linear_attn.in_proj_qkv",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 8192,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.linear_attn.in_proj_z",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 4096,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.linear_attn.out_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 4096,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 256,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.shared_expert.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 2048,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.shared_expert.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.shared_expert.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 512,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.shared_expert_gate",
"mode": "affine",
"bits": 8,
"group_size": 64,
"rows_checked": 1,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.switch_mlp.down_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 524288,
"columns": 512,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.switch_mlp.gate_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
},
{
"module": "language_model.model.layers.9.mlp.switch_mlp.up_proj",
"mode": "affine",
"bits": 5,
"group_size": 64,
"rows_checked": 131072,
"columns": 2048,
"all_dequantized_values_finite": true
}
],
"all_dequantized_values_finite": true,
"tokenizer_source_files": {
"tokenizer.json": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523",
"tokenizer_config.json": "792fa3f0cb88b111e54ef3134c873531008c4df471d108da17903426e308aa7b",
"chat_template.jinja": "e84f32a23fdda27689f868aa4a1a5621f41133e51a48d7f3efcbea2839574259"
},
"functional_passed": 8,
"functional_total": 8,
"all_generation_logits_finite": true
}